Our overall goal is to develop an informatics infrastructure to support basic cancer genetics research using the mouse as a model system of human cancer. We have developed the Mouse Tumor Biology Database (MTB; http://tumor.informatics.jax.org) as a prototype to explore ways to effectively support electronic access to mouse tumor data. We now propose to build on this resource, maturing it from a prototype database into a robust database system, expanding the depth and breadth of data coverage, enhancing the collection of tumor images, and improving search capabilities and graphical displays. We will integrate data on frequency, incidence, genetics, and pathology of neoplastic disorders, emphasizing the acquisition of data on tumors that develop characteristically in different genetically defined strains of mice (inbred, mutant, and genetically engineered). To accomplish our goal we will: migrate MTB to a robust relational database management system develop new user interfaces, query capabilities, and graphical displays continue to populate MTB with data on strain-specific patterns of tumorigenesis expand data coverage to include cancer susceptibility/severity modifier genes and QTL enhance and expand pathology-related data (text and images) develop database structures to accommodate in-vivo imaging techniques build on collaborative efforts with veterinary pathologists work with the Mouse Models of Human Cancer Consortium (MMHCC) to facilitate data sharing increase the number of links between MTB and relevant, cancer- related resources, particularly those capturing gene expression profiling that will further refine the categorization of tumor subtypes.